{"slug":"content-humanizer-principles","title":"Content Humanizer: 7 Principles + Scoring","tags":["humanizing","voice","writing-quality","ai-detection"],"agent_summary":"Seven editing principles to make AI-generated content sound human: specificity, personality injection, corporate-speak removal, sentence variation, conversational devices, and actionable lists. Includes composite scoring formula.","trigger_phrases":["humanize content","make content sound human","remove ai writing","content sounds robotic","add personality to content","pass ai detector"],"runnable":true,"markdown":"\n# Content Humanizer: 7 Principles + Scoring\n\nRun **after** `/content-scrubber`. Load brand docs before applying if voice-matching a specific client.\n\n## 5-Step Analysis First\n\nRun in order before editing:\n\n1. **Humanity Check** — Does it sound like a real person? Are there opinions, takes, personal observations?\n2. **Readability** — Sentence length variation? Paragraph max 4 sentences?\n3. **Personality/Voice** — Detectable POV? Tone matches brand?\n4. **Specificity/Examples** — Claims backed by specific numbers, names, scenarios?\n5. **Engagement/Flow** — Does the opening compel reading past line 3?\n\n---\n\n## Robotic vs Human Signals\n\n| Robotic Red Flags | Human Green Flags |\n|---|---|\n| \"In today's fast-paced world...\" | Opens with specific scenario or provocative claim |\n| \"It's important to note that...\" | Direct statement — just say the thing |\n| \"Furthermore,\" / \"Moreover,\" | \"Also.\" / \"Here's the thing.\" / \"And.\" |\n| Lists of 10+ generic tips | 3-5 tips with real examples |\n| Every paragraph same length | Rhythm varies — one line paragraphs exist |\n| No opinions, just facts | \"In my experience...\" / \"Most people get this wrong.\" |\n| Benefit-free CTAs (\"Learn more\") | Outcome-driven CTAs (\"Get your first result in 24 hours\") |\n| Generic examples (\"a business owner\") | Specific examples (\"a 12-person SaaS team\") |\n\n---\n\n## 7 Editing Principles\n\n### 1. Show, Don't Tell\n**Before:** \"Our product is incredibly effective at boosting productivity.\"\n**After:** \"Teams using this cut their weekly reporting time from 4 hours to 40 minutes — in week one.\"\n\n### 2. Inject Personality\nAdd opinions, admits, and takes. Replace hedge-language with conviction.\n**Before:** \"There are various approaches you might consider.\"\n**After:** \"Most people try X first. It works, but it's slow. Here's the faster path.\"\n\n### 3. Kill Corporate Speak\nBan: leverage, utilize, synergy, robust, holistic, seamless, cutting-edge, best-in-class, game-changing. Replace with plain Anglo-Saxon equivalents.\n\n### 4. Add Specific Details\nEvery major point needs a specific anchor: a number, a name, a scenario, a timeframe.\n\nQuality spectrum:\n- Vague: \"Our customers see results.\"\n- Generic: \"Customers report improved productivity.\"\n- Specific: \"87% of customers report saving 5+ hours per week.\"\n- Compelling: \"Sarah, a 3-person agency owner, cut her proposal time from 2 hours to 18 minutes.\"\n\n### 5. Vary Sentence Structure\nShort. Punchy. Then a longer one that builds context and earns the period. Then a question — why not? One-word sentences exist. So do fragments.\n\n### 6. Use Conversational Devices\n- Direct reader address: \"You already know this, but...\"\n- Rhetorical questions: \"Sound familiar?\"\n- Admissions: \"Here's what most guides won't tell you.\"\n- Callbacks: Reference something mentioned earlier.\n\n### 7. Make Lists Actionable\nEvery bullet must complete a task or give a specific insight. No orphan bullets that just name a concept without explaining it.\n\n---\n\n## Engagement Requirements\n\n- **Hook types** (one per article): Provocative claim, surprising stat, relatable pain point, counter-intuitive take, or specific scenario\n- **Mini-stories**: 2-3 per article minimum — named character, situation, outcome\n- **CTA distribution**: Soft CTA mid-article, hard CTA at close\n- **Paragraph max**: 4 sentences\n\n---\n\n## Composite Score Formula\n\n| Dimension | Weight |\n|---|---|\n| Humanity | 30% |\n| Specificity | 25% |\n| Structure balance | 20% |\n| SEO | 15% |\n| Readability | 10% |\n\n**Pass threshold: ≥ 70 composite score**\n\n`(humanity × 0.30) + (specificity × 0.25) + (structure_balance × 0.20) + (seo × 0.15) + (readability × 0.10)`\n\n---\n\n## Self-Check Before Submitting\n\n- [ ] Would I stop scrolling to read this?\n- [ ] Can I identify whose voice this is?\n- [ ] Is every major claim supported by something specific?\n- [ ] Are there at least 2 moments where a human personality is visible?\n- [ ] Would this pass a \"did a real person write this?\" gut check?\n\n## Red Lines\n\n- No fabricated statistics\n- No false testimonials or invented names presented as real\n- Never strip accurate information to sound simpler\n\n#content-sop #humanizing #voice\n","html":"<h1>Content Humanizer: 7 Principles + Scoring</h1>\n<p>Run <strong>after</strong> <code>/content-scrubber</code>. Load brand docs before applying if voice-matching a specific client.</p>\n<h2>5-Step Analysis First</h2>\n<p>Run in order before editing:</p>\n<ol>\n<li><strong>Humanity Check</strong> — Does it sound like a real person? Are there opinions, takes, personal observations?</li>\n<li><strong>Readability</strong> — Sentence length variation? Paragraph max 4 sentences?</li>\n<li><strong>Personality/Voice</strong> — Detectable POV? Tone matches brand?</li>\n<li><strong>Specificity/Examples</strong> — Claims backed by specific numbers, names, scenarios?</li>\n<li><strong>Engagement/Flow</strong> — Does the opening compel reading past line 3?</li>\n</ol>\n<hr>\n<h2>Robotic vs Human Signals</h2>\n<p>| Robotic Red Flags | Human Green Flags |\n|---|---|\n| \"In today's fast-paced world...\" | Opens with specific scenario or provocative claim |\n| \"It's important to note that...\" | Direct statement — just say the thing |\n| \"Furthermore,\" / \"Moreover,\" | \"Also.\" / \"Here's the thing.\" / \"And.\" |\n| Lists of 10+ generic tips | 3-5 tips with real examples |\n| Every paragraph same length | Rhythm varies — one line paragraphs exist |\n| No opinions, just facts | \"In my experience...\" / \"Most people get this wrong.\" |\n| Benefit-free CTAs (\"Learn more\") | Outcome-driven CTAs (\"Get your first result in 24 hours\") |\n| Generic examples (\"a business owner\") | Specific examples (\"a 12-person SaaS team\") |</p>\n<hr>\n<h2>7 Editing Principles</h2>\n<h3>1. Show, Don't Tell</h3>\n<p><strong>Before:</strong> \"Our product is incredibly effective at boosting productivity.\"\n<strong>After:</strong> \"Teams using this cut their weekly reporting time from 4 hours to 40 minutes — in week one.\"</p>\n<h3>2. Inject Personality</h3>\n<p>Add opinions, admits, and takes. Replace hedge-language with conviction.\n<strong>Before:</strong> \"There are various approaches you might consider.\"\n<strong>After:</strong> \"Most people try X first. It works, but it's slow. Here's the faster path.\"</p>\n<h3>3. Kill Corporate Speak</h3>\n<p>Ban: leverage, utilize, synergy, robust, holistic, seamless, cutting-edge, best-in-class, game-changing. Replace with plain Anglo-Saxon equivalents.</p>\n<h3>4. Add Specific Details</h3>\n<p>Every major point needs a specific anchor: a number, a name, a scenario, a timeframe.</p>\n<p>Quality spectrum:</p>\n<ul>\n<li>Vague: \"Our customers see results.\"</li>\n<li>Generic: \"Customers report improved productivity.\"</li>\n<li>Specific: \"87% of customers report saving 5+ hours per week.\"</li>\n<li>Compelling: \"Sarah, a 3-person agency owner, cut her proposal time from 2 hours to 18 minutes.\"</li>\n</ul>\n<h3>5. Vary Sentence Structure</h3>\n<p>Short. Punchy. Then a longer one that builds context and earns the period. Then a question — why not? One-word sentences exist. So do fragments.</p>\n<h3>6. Use Conversational Devices</h3>\n<ul>\n<li>Direct reader address: \"You already know this, but...\"</li>\n<li>Rhetorical questions: \"Sound familiar?\"</li>\n<li>Admissions: \"Here's what most guides won't tell you.\"</li>\n<li>Callbacks: Reference something mentioned earlier.</li>\n</ul>\n<h3>7. Make Lists Actionable</h3>\n<p>Every bullet must complete a task or give a specific insight. No orphan bullets that just name a concept without explaining it.</p>\n<hr>\n<h2>Engagement Requirements</h2>\n<ul>\n<li><strong>Hook types</strong> (one per article): Provocative claim, surprising stat, relatable pain point, counter-intuitive take, or specific scenario</li>\n<li><strong>Mini-stories</strong>: 2-3 per article minimum — named character, situation, outcome</li>\n<li><strong>CTA distribution</strong>: Soft CTA mid-article, hard CTA at close</li>\n<li><strong>Paragraph max</strong>: 4 sentences</li>\n</ul>\n<hr>\n<h2>Composite Score Formula</h2>\n<p>| Dimension | Weight |\n|---|---|\n| Humanity | 30% |\n| Specificity | 25% |\n| Structure balance | 20% |\n| SEO | 15% |\n| Readability | 10% |</p>\n<p><strong>Pass threshold: ≥ 70 composite score</strong></p>\n<p><code>(humanity × 0.30) + (specificity × 0.25) + (structure_balance × 0.20) + (seo × 0.15) + (readability × 0.10)</code></p>\n<hr>\n<h2>Self-Check Before Submitting</h2>\n<ul>\n<li>[ ] Would I stop scrolling to read this?</li>\n<li>[ ] Can I identify whose voice this is?</li>\n<li>[ ] Is every major claim supported by something specific?</li>\n<li>[ ] Are there at least 2 moments where a human personality is visible?</li>\n<li>[ ] Would this pass a \"did a real person write this?\" gut check?</li>\n</ul>\n<h2>Red Lines</h2>\n<ul>\n<li>No fabricated statistics</li>\n<li>No false testimonials or invented names presented as real</li>\n<li>Never strip accurate information to sound simpler</li>\n</ul>\n<p>#content-sop #humanizing #voice</p>\n"}